OBJECTIVE: To investigate the relationship between mononuclear cell infiltration of primary ovarian cancer and the subsequent survival of the patient. Although prognosis is reported to be better when a mononuclear infiltrate is present, there has never been an investigation of the nature, function and potential role of these inflammatory cells. The question of whether or not this reaction is immunologic in nature or simply a non-specific reaction to tumor cell death has not been answered. We have developed a methodology which has been exploited to investigate the relationship between systemic and in situ tumor immunity in a wide variety of animal models. It seemed important to extend these studies to the clinic and, in particular, to ovarian cancer. These patients have generally poor survival rates and large tumor burdens, and they routinely produce large quantities of malignant effusions. In spite of this, many patients show remarkable response to chemotherapy and there are wide variations in the proportions of inflammatory cells and tumor cells recoverable from their tumors and effusions. Our initial investigations have concentrated on isolating the infiltrating cells present in these tumors and characterizing the various fractions by a variety of membrane and cytoplasmic markers. We have initiated a study of the functional capacity of these cells in a series of tests for non-specific immune reactivity - PHA, NK and ADCC. Our future studies will concentrate on the activity these cells have in relation to the autologous tumor cells both in preparations derived from solid tumors and in ascites.